Risk prediction for suicide in schizophrenia is known to be complex. Illness-related risk factors are important predictors, with number of prior suicide attempts, depressive symptoms, active hallucinations and delusions.

To analyze the impact of clinical features of schizophrenia on suicidal behaviour and analyze relationship between violence and suicidality.

We compared a group of 28 suicidal, 32 hetero-aggressive and 60 non-aggressive male patients with schizophrenia. The severity of the psychopathology was assessed using PANSS, hetero-aggressiveness using OAS and suicidality using C-SSRS and InterSePT scales.

Suicidality is in positive correlation with the presence of depression (r = 0.485, P < 0.001) and imperative hallucinations (r = 0.391, P < 0.001) but in negative correlation with total PANSS score (r = –0.297, P < 0.001), impulsiveness (r = –0,237, P < 0.001) and hetero-aggressive behaviour (r = –0.551, P < 0.001). Previous attempt was found in 53% suicidal patients (P = 0.047). Heteroaggressiveness is in positive correlation with total PANSS (r = 0.43, P < 0.001) and impulsive behaviour (r = 0.57, P < 0.001). This study supports the contention that high impulsivity in schizophrenia patients is significant in the etiology of violent but not suicidal behaviour.

In the order to diminish the level of suicide among patients with schizophrenia we must be aware of the importance of depressive symptoms, imperative hallucinations and previous attempts. This study supports the contention that high impulsivity in schizophrenia patients is significant in the etiology of violence but not suicidal behaviour.

The authors have not supplied their declaration of competing interest.